Does anyone know of any paper where the classical Smagorinsky's LES model is compared with the dynamic Smagorinsky's LES model (Germano)?

Hi,

Vreman et al. (1997), J. Fluid Mech. 339, pp. 357-390. In this paper both models are compared I think.

Tom

Thanks Tom, this is exactly what I was looking for. Do you have any advice on how I can quantitatively compare data from 2 simulations using different LES models?

Hi Jessica,

I have no personal experience in testing LES models, so I can give only some very global advice. Hopefully, you have some DNS or experimental data you can compare with. Then you can perform a priori tests for example: filter the DNS data and see how well the predictions of the LES models for the subgrid scales based on the fitered DNS data correlate with the subgrid scales in the DNS. In this case you don't have to perform simulations. I have read somewhere that the Smagorinsky model gives quite bad results in the a priori tests, thus low correlations between the subgrid scales in the LES and DNS. Another possibility are a posteriori tests: perform simulations with the LES models and compare these with DNS or experimental data. Mostly mean velocity profiles and rms profiles are compared. Other possiblities are correlation lenghts for small and large scales, and spectra. A LES model that performs better in the a priori tests than another does not need to give better results in the a posteriori tests. Moreover, some models destabilize the simulations. A posteriori tests are thus always meaningful.

Tom